Michele Rienzner

Modeling water resources of a highly irrigated alluvial plain (Italy): calibrating soil and groundwater models

Modern and effective water management in large alluvial plains that have intensive agricultural activity requires the integrated modeling of soil and groundwater. The models should be complex enough to properly simulate several, often non-linear, processes, but simple enough to be effectively calibrated with the available data. An operative, practical approach to calibration is proposed, based on three main aspects. First, the coupling of two models built on well-validated algorithms, to simulate (1) the irrigation system and the soil water balance in the unsaturated zone and (2) the groundwater flow. Second, the solution of the inverse problem of groundwater hydrology with the comparison model method to calibrate the model. Third, the use of appropriate criteria and cross-checks (comparison of the calibration results and of the model outputs with hydraulic and hydrogeological data) to choose the final parameter sets that warrant the physical coherence of the model. The approach has been tested by application to a large and intensively irrigated alluvial basin in northern Italy.RésuméLa gestion moderne et efficace de l’eau dans les grandes plaines alluviales siège d’une agriculture intensive requiert la modélisation intégrée du sol et des eaux souterraines. Les modèles doivent être suffisamment complexes pour simuler correctement plusieurs processus souvent non linéaires, mais suffisamment simples pour être calibrés efficacement avec les données disponibles. Une approche de calibration pratique et opérationnelle est proposée, basée sur trois aspects principaux. Premièrement, le couplage de deux modèles construits sur des algorithmes bien validés, pour simuler (1) le dispositif d’irrigation et le bilan en eau du sol au sein de la zone non saturée et (2) l’écoulement d’eau souterraine. Deuxièmement, la résolution du problème hydrogéologique inverse avec la Méthode de Comparaison de Modèle pour caler le modèle. Troisièmement, l’utilisation de critères appropriés et de vérifications croisées (comparaison des résultats du calage et des sorties du modèle avec des données hydrologiques et hydrogéologiques) pour choisir les jeux de paramètres finaux qui garantissent la cohérence physique du modèle. L’approche a été testée par application à un vaste bassin alluvial intensément irrigué du Nord de l’Italie.ResumenEl manejo moderno y efectivo del agua en grandes planicies aluviales que tienen una intensa actividad agrícola requiere el modelado integrado del suelo y del agua subterránea. Los modelos deber ser lo suficientemente complejos como para simular correctamente varios procesos, a menudo no lineales, pero lo suficientemente simple para ser efectivamente calibrado con los datos disponibles. Se propone un enfoque práctico y operativo de la calibración, basada en tres aspectos principales. Primero, el acoplamiento de dos modelos construidos sobre algoritmos bien validados, para simular (1) el sistema de irrigación y el balance de agua en el suelo en la zona no saturada y (2) el flujo de agua subterránea. Segundo, la solución del problema inverso de hidrología de agua subterránea con el Método de comparación de modelos para calibrar el modelo. Tercero, el uso de criterios apropiados y controles cruzados (comparación entre los resultados de la calibración y las salidas del modelo con los datos hidrogeológicos e hidráulicos) para elegir el conjunto de parámetros finales que garanticen la coherencia física del modelo. El enfoque ha sido probado por su aplicación a una gran cuenca aluvial, intensamente irrigada en el norte de Italia.摘要在有密集型农业活动的大型冲积平原进行有效的现代水资源管理需要对土壤和地下水进行综合模拟。模型需要足够复杂,以模拟几种常见的非线性的过程,同时又要足够简单,能够采用现有数据进行校准。本文主要基于以下三个方面,提出了一个具备可操作性且符合实际的校准方法:1. 基于有效算法的两种模型的耦合,以模拟(1)非饱和带灌溉系统和土壤水平衡和(2)地下水流。2. 针对地下水文学的反演问题,采用比较法去校正模型。3.使用合适的准则和交互检验(根据水力和水文地质数据,比较模型的校准结果和模型输出结果)去选择最后的参数设置以确保模型的物理一致性。将该方法已经应用在意大利北部的一个大型密集灌溉冲积平原中进行ResumoUma moderna e eficaz gestão da água nas grandes planícies aluviais com atividade agrícola intensiva requer uma modelação integrada do solo e das águas subterrâneas. Os modelos devem ser suficientemente complexos para simular adequadamente vários processos, muitas vezes de caraterísticas não-lineares, mas suficientemente simples para serem adequadamente calibrados com os dados disponíveis. Neste artigo é proposta uma abordagem para calibração baseada em três aspetos principais. Primeiro, o acoplamento de dois modelos construídos sobre algoritmos bem validados, para simular (1) o sistema de rega e balanço hídrico do solo na zona não saturada e (2) o fluxo de águas subterrâneas. Segundo, uma solução do problema inverso da hidrologia subterrânea, baseada no Método de Comparação do Modelo para calibrar o modelo. Terceiro, o uso de critérios adequados e validações cruzadas (comparação dos resultados de calibração com as saídas do modelo com dados hidráulicos e hidrogeológicos), de modo a selecionar o conjunto final de parâmetros que garantem a coerência física do modelo. A abordagem tem sido testada numa bacia aluvial extensa e intensamente sujeita a rega, localizada no norte da Itália.

An integrated, multisensor system for the continuous monitoring of water dynamics in rice fields under different irrigation regimes

The cultivation of rice, one of the most important staple crops worldwide, has very high water requirements. A variety of irrigation practices are applied, whose pros and cons, both in terms of water productivity and of their effects on the environment, are not completely understood yet. The continuous monitoring of irrigation and rainfall inputs, as well as of soil water dynamics, is a very important factor in the analysis of these practices. At the same time, however, it represents a challenging and costly task because of the complexity of the processes involved, of the difference in nature and magnitude of the driving variables and of the high variety of field conditions. In this paper, we present the prototype of an integrated, multisensor system for the continuous monitoring of water dynamics in rice fields under different irrigation regimes. The system consists of the following: (1) flow measurement devices for the monitoring of irrigation supply and tailwater drainage; (2) piezometers for groundwater level monitoring; (3) level gauges for monitoring the flooding depth; (4) multilevel tensiometers and moisture sensor clusters to monitor soil water status; (5) eddy covariance station for the estimation of evapotranspiration fluxes and (6) wireless transmission devices and software interface for data transfer, storage and control from remote computer. The system is modular and it is replicable in different field conditions. It was successfully applied over a 2-year period in three experimental plots in Northern Italy, each one with a different water management strategy. In the paper, we present information concerning the different instruments selected, their interconnections and their integration in a common remote control scheme. We also provide considerations and figures on the material and labour costs of the installation and management of the system.

Critical values improvement for the standard normal homogeneity test by combining Monte Carlo and regression approaches

ABSTRACT The distribution of the test statistics of homogeneity tests is often unknown, requiring the estimation of the critical values through Monte Carlo (MC) simulations. The computation of the critical values at low α, especially when the distribution of the statistics changes with the series length (sample cardinality), requires a considerable number of simulations to achieve a reasonable precision of the estimates (i.e. 106 simulations or more for each series length). If, in addition, the test requires a noteworthy computational effort, the estimation of the critical values may need unacceptably long runtimes. To overcome the problem, the paper proposes a regression-based refinement of an initial MC estimate of the critical values, also allowing an approximation of the achieved improvement. Moreover, the paper presents an application of the method to two tests: SNHT (standard normal homogeneity test, widely used in climatology), and SNH2T (a version of SNHT showing a squared numerical complexity). For both, the paper reports the critical values for α ranging between 0.1 and 0.0001 (useful for the p-value estimation), and the series length ranging from 10 (widely adopted size in climatological change-point detection literature) to 70,000 elements (nearly the length of a daily data time series 200 years long), estimated with coefficients of variation within 0.22%. For SNHT, a comparison of our results with approximated, theoretically derived, critical values is also performed; we suggest adopting those values for the series exceeding 70,000 elements.